GNN and TGNN Performance: Benchmark Results for Temporal Graph Neural Networks

Performance results for Graph Neural Networks (GNNs) and Temporal Graph Neural Networks (TGNNs), including the DoomBot benchmark. The dataset was authored by Marcus Araujo and last updated on April 22, 2026. It is a small 5.5 KB Excel file containing tabular results.

Use Cases

• Benchmarking Temporal GNN (TGNN) architectures based on the included DoomBot benchmark.
• Analyzing the impact of hyperparameters like temporal window (W) and learning rate (LR) on model performance.
• Comparing optimizer (Opt) and architectural choices like hidden channels (HC) and layers (#L) across experiments.
• Studying regularization effects using metrics for dropout (Drop) and weight decay (WD).

Strengths

• Dataset is licensed under CC-BY-4.0, permitting open sharing and adaptation.
• Specific hyperparameters and metrics are defined, including temporal window (W), hidden channels (HC), and p-values.

Limitations

• Row count is unknown, which may limit suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 5.5 KB, indicating limited scope.

Provenance

Source

figshare

Freshness

Last updated 2026-04-22 17:44:25; freshness should be verified.